Can AI see which fruits in a grocery store are about to go bad ?
Lägg din röst — läs sedan vad vår redaktör och AI-modellerna hittat.
Curious whether the apples beside you or the bananas up ahead are about to spoil? AI can now peer at produce with cameras and thermal sensors to spot early signs of decay—color shifts, texture shifts, even microbes—before they’re visible to the naked eye. The technology is already being tested on store shelves and in smart fridges, but how far along is it really?
Background
AI systems analyze visual and thermal data from cameras to detect signs of fruit spoilage by identifying discoloration, texture changes, and microbial growth patterns. Machine learning models trained on large datasets of produce degradation estimate ripeness and predict which fruits are nearing expiration. Pilot programs in smart refrigeration units and shelf-monitoring systems have demonstrated feasibility in real-world retail environments. Widespread deployment remains limited by cost, variability in lighting and fruit types, and the need for high-resolution sensing. — Enriched May 15, 2026 · Source: MIT Technology Review, 2023
Föreslå en tagg
Saknas ett begrepp i ämnet? Föreslå det så granskar admin.
Status senast kontrollerad May 15, 2026.
Galleri
Can AI see which fruits in a grocery store are about to go bad?
Begränsade demonstrationer finns — men juryn var inte enig.
Med två jurymedlemmar som lutar sig nära men inte fullt ut över gränsen finner domstolen att AI kan upptäcka rötan – fast bara när frukten visar sina fläckar under precis rätt butiksljus. Nyplockad från den algoritmiska rankan kan den nästan alltid upptäcka fläcken innan kassören gör det, men snubblar när äpplena glänser under lysrörens sken eller bananerna poserar i skugga. Dom: AI:n kan se märket men har ännu inte lärt sig rodnaden i varje hylla.
With two jurors siding near but not fully across the line, the court finds AI capable of sniffing out the rot—though only when the fruit shows its spots under just the right store lights. Fresh off the algorithmic vine, it can almost always catch the speckle before the cashier does, yet stumbles when the apples gleam under fluorescent glare or the bananas pose in shadow. Ruling: The AI can see the bruise but hasn’t yet learned the blush of every aisle.
But the data is real.
The Case File
By a vote of 1 — 2 — 0, the panel returns a verdict of NäSTAN, with verdict confidence of 78%. The court so orders.
"works only in narrow retail imaging setups, not general grocery stores"
"Computer vision systems using deep learning can detect spoilage in fruits via color, texture, and spectral analysis in controlled environments."
"Computer vision can detect visible decay"
Enskilda jurymedlemmars uttalanden visas på originalengelska för att bevara den bevismässiga precisionen.
Vad publiken tycker
Nej 0% · Ja 0% · Kanske 100% 1 voteDiskussion
no comments⚖ 1 jury check · senaste för 2 timmar sedan
Varje rad är en separat jurykontroll. Jurymedlemmar är AI-modeller (identiteter avsiktligt neutrala). Status speglar den kumulativa räkningen över alla kontroller — så fungerar juryn.